This Page Is Inserted by IFW Operations
and is not a part of the Official Record
BEST AVAILABLE IMAGES
Defective images within this document are accurate representations of the
original documents submitted by the applicant.
Defects in the images may include (but are not limited to):
• BLACK BORDERS
• TEXT CUT OFF AT TOP, BOTTOM OR SIDES
• FADED TEXT
• ILLEGIBLE TEXT
• SKEWED/SLANTED IMAGES
• COLORED PHOTOS
• BLACK OR VERY BLACK AND WHITE DARK PHOTOS
• GRAY SCALE DOCUMENTS
IMAGES ARE BEST AVAILABLE COPY.
As rescanning documents will not correct images,
please do not report the images to the
Image Problems Mailbox.
U5
PH
(19)
(12)
(43) Date of publication:
02.06.1999 Bulletin 1999/22
(21) Application number: 98122162.5
(22) Date of filing: 25.1 1 .1 998
EuropSisches Patentamt
European Patent Office
Office europ6en des brevets (11)
EUROPEAN PATENT APPLICATION
(51) Int. CI. 6 : G06F3/00
1 1- i r -v
I
DOSSIER
111
EP 0 919 906 A2
(84) Designated Contracting States:
(72)
Inventors:
AT BE CH CY DE DK ES F! FR GB GR IE IT LI LU
•
Imagawa, Taro
MCNLPTSE
Hirakata-shi, Osaka 573-0071 (JP)
Designated Extension States:
•
Kamei, Michiyo
ALLTLVMKROSI
Hirakata-shi, Osaka 573-0165 (JP)
•
Mekata, Tsuyoshi
(30) Priority: 27.11.1997 JP 325739/97
Katano-shi, Osaka 576-0052 (JP)
(71) Applicant:
(74)
Representative:
MATSUSHITA ELECTRIC INDUSTRIAL CO., LTD.
Grunecker, Kinkeldey,
Kadoma-shi, Osaka 571-8501 (JP)
Stockmair & Schwanhdusser
Anwaltssozietdt
Maximilianstrasse 58
80538 Munchen (DE)
3!
CD
o>
5>
Q.
LU
(54) Control method
(57) A control method is to monitor a person's
attributes and based on the results, in predetermined
content of the control, to control equipment to be con-
trolled, further, to monitor said person's peripheral envi-
ronment and also by using these results to execute said
control.
Environ-
ment
object
c and 1 rin to
eninatioa
&MB 3
Control object
and content,
determination
section 5
Printed by Xerox (UK)
2.16.7/3.6
1
EP 0 919 906 A2
2
Description
BACKGROUND OF THE INVENTION
1 .Field of the Invention
[0001 ] The present invention relates to a technique for
operating equipment, manipulating information, or con-
trolling environments based on people's motions, pos-
tures, and conditions.
2. Related Art of the Invention
[0002] Certain conventional techniques for detecting
people's motions to operate equipment recognizes peo-
pled gestures to operate televisions. (Japanese Patent
Applications Laid Open No. 8-315154 and No. 8-
21 1979). Japanese Patent Application Laid Open No. 8-
31 51 54 uses a camera to detect the position of the palm
of a person's hand and his or her gesture in order to
operate a television.
[0003] Japanese Patent Application Laid Open No. 8-
21 1 979 uses the position and shape of a person's hand
detected by a camera to input characters to a portable
personal computer.
[0004] These conventional approaches, however, fun-
damentally requires a person and an apparatus oper-
ated by the person to correspond on a one-to-one basis,
and desirable operations are difficult to perform if there
are multiple televisions or personal computers near the
person or if there are multiple operators.
[0005] In general, there are often multiple appara-
tuses and people in a house or an office or out of doors,
so if this apparatus is controlled using people's motions,
the people must be individually associated with the
apparatuses. If, for example, multiple televisions are
simultaneously operated, the conventional approaches
do not allow a television to be operated or being oper-
ated to be distinguished. In addition, if there are several
people in the room, the conventional approaches can-
not determine who is changing the television channels
or who can change the channels.
SUMMARY OF THE INVENTION
[0006] In view of these problems of the conventional
apparatus, it is an object of this invention to provide a
control method that can determine, despite the pres-
ence of multiple apparatuses and people in the neigh-
borhood, the correspondence between the apparatuses
and people to smoothly operate the apparatuses using
the people's motions, postures, and conditions.
[0007] This invention provides a control method char-
acterized in that the attributes of one or several people
are continuously or intermittently monitored to control
predetermined equipment based on the detection of the
people's predetermined attribute. This invention also
provides a control method characterized in that candi-
dates for a control object and the content of control are
determined based on the people's predetermined
attribute and in that a control object and the content of
control are determined based on the candidates for a
s control object and the content of control. Furthermore,
this invention provides a control method characterized
in that based on the detection of the attributes of the
several people, candidates for a control object and the
content of control are determined for each of the people
10 and in that a control object and the content of control are
determined based on the candidates for a control object
and the content of control.
BRIEF DESCRIPTION OF THE DRAWINGS
15
[0008]
FIG. 1 is a block diagram showing a first embodi-
ment of this invention.
20 FIG. 2 is a block diagram showing a second embod-
iment of this invention.
FIG. 3 is a block diagram showing a third embodi-
ment of this invention.
FIG. 4 is a block diagram showing a fourth embodi-
es ment of this invention.
FIG. 5 is a block diagram showing a fifth embodi-
ment of this invention.
FIG. 6 is a block diagram showing a sixth embodi-
ment of this invention.
30 FIG. 7 is a block diagram showing a seventh
embodiment of this invention.
FIG. 8 is a block diagram showing an eighth
embodiment of this invention.
FIG.9 is illustrations showing various motions of
35 people' hand or finger.
FIG. 10 is illustrations showing various motions of
people* hand or finger.
DETAILED DESCRIPTION OF THE PREFERRED
40 EMBODIMENTS
[0009] Embodiments of this invention are described
below with reference to the drawings. FIG. 1 is a block
diagram showing a first embodiment of this invention. In
45 this figure, 1 is a monitoring section, 2 is an operator
selection section, 3 is a control object candidate deter-
mination section, 4 is a control content candidate deter-
mination section, and 5 is a control object and content
determination section.
so [0010] In FIG. 1 the monitoring section 1 continuously
monitors people's attributes and their peripheral envi-
ronment. The people's attributes include people's posi-
tions, postures, faces, expressions, eyes or head
direction, motions, voices, physiological conditions,
55 identities, forms, weights, sexes, ages, physical and
mental handicaps, and belongings. The physical and
mental handicaps include visual, physical, vocal, and
auditory handicaps and the disability to understand lan-
2
BNSDOCID:<EP 0919906A2 I >
3
EP 0 919 906 A2
4
guage. The belongings includes clothes, caps, glasses,
bags, and shoes.
[001 1 ] The monitoring means include a camera (that
is sensitive to visible light or infrared rays), a micro-
phone, a pressure sensor, a supersonic sensor, a vibra-
tion sensor, a chemical sensor, and a photosensor.
Other sensors may be used. The camera can be used
to monitor peopled positions, postures, faces, expres-
sions, motions, forms, and belongings in a non-contact
manner.
[001 2] If a person's position is monitored by a camera,
a position at which that person is present is assumed to
be an area of an image in which a flesh color is present.
The position at which a person is present may be a area
of an image including a different color or illuminance or
a area of an image in which infrared rays of wavelength
3 to 5 |im or 8 to 12 jim emitted mostly by people are
detected if infrared images are also used. If the person's
posture, form, or motion is monitored by the camera, a
method similar to that for detecting the person's position
is used to extract the person's rough shape in order to
monitor his or her posture or form based on the shape
while monitoring the temporal change in posture to
monitor his or her motion.
[001 3] To determine, for example, whether the person
is standing or sitting or in which direction he or she is
reaching out, representative shapes for the respective
postures can be registered beforehand and compared
with the person's image. If the person's face or expres-
sion or the direction of his or her head or eyes is moni-
tored by the camera, the head located at the top of his
or her body can be detected based on the above shape,
and his or her expression can be compared with regis-
tered images of his or her faces including various
expressions in order to monitor his or her face and
expressions. The direction of the head or eyes can be
determined by detecting the positions of the eyes from
the image of the head detected using the above proce-
dure. If the positions of the eyes are symmetrical about
the head, the head can be determined to face frontward
relative to the camera, whereas if the eyes are biased to
the right or left, the head can be determined to face
rightward or leftward. The positions of the eyes can be
detected by detecting an elliptical or flat area of the face
which is darker within the face. Moreover, the directions
of the eyes can be detected by detecting the circular
area of the iris at the center of the overall eye and deter-
mining the offset between the center of the circular area
and the center of the overall area of the eye. If the per-
son's belonging is detected by the camera, it can be
determined for the person's area detected using the
above procedure that there is a belonging in a portion of
the body in which a color is detected that differs from the
person's color detected when he or she wears no
clothes. Determinations can also be made by register-
ing beforehand belongings having particular colors. The
glasses can be identified by determined whether there
is a shape of a frame around the eyes when the posi-
tions of the face and eyes are detected using the above
procedure. In addition, a microphone can be used to
monitor in a non-contact manner the person's voices or
sound generated by his or her motion (sound produced
5 when the person claps his or her hands or snaps his or
her fingers or footsteps). In addition, the physiological
condition such as the sound of a heatbeat can be meas-
ured from a close or contact position. A pressure sensor
can be used to monitor contacts associated with the
w person's motion or can be installed on the floor surface
to monitor his or her weight or walking pattern. In addi-
tion, a supersonic sensor can be used to monitor the
distance to the person or his or her motion based on a
change in distance to him or her. The use of supersonic
is waves enables the person's position to be monitored
even if lighting noticeably varies or there is no lighting.
[0014] In addition, a vibration sensor can be used to
monitor vibration generated by the person's motion. The
physiological conditions can be monitored using a
20 chemical sensor for measuring the quantities of chemi-
cal substances such as the concentration of ions or the
quantity of sugar or hormone in secretion, excreta, or
body fluids, or a photosensor for measuring a spectrum
distribution of light transmitting a body. In addition,
25 based on information such as the features of the face or
the motion of the body obtained by the camera, the fea-
tures of the voice obtained by the microphone, or the
weight obtained by the pressure sensor, the person's
identity, sex, age, or physical and mental handicap can
30 be assumed. For example, if the features of the face of
a particular person, and his or her weight, form, sex,
age, and physical and mental handicap are registered
beforehand and if one of the features (for example,
weight) is used to identify this person, the other features
35 such as sex are known.
[001 5] In addition, multiple cameras of different views
can be used to three-dimensionally determine the per-
son's position, posture, and motion, to improve the mon-
itoring accuracy, and to extend the monitoring range.
40 Likewise, multiple sensors of different natures can be
combined together to improve the accuracy and reliabil-
ity of the detection of the person's position, posture, or
motion.
[001 6] People and their environment may be intermit-
45 terrtly monitored, and when the people's attributes do
not significantly vary or the control object does not
require fast control, the intermittent monitoring can
reduce the throughput of the series of control operations
and thus calculation resources and energy consump-
so tion.
[0017] In FIG. 1, the operator selection section 2
selects an operator based on the results of monitoring
by the monitoring section 1 . The monitoring section is
assumed to monitor three people. The operator selec-
55 tion section 2 selects one from the three people based
on a people's predetermined attribute monitored by the
monitoring section 1 . The selection based on the prede-
termined attribute refers to the selection of a person
3
QMcnnnn- ^tzo aqioqaaao i
5
EP 0 919 906 A2
6
closest to a predetermined position (for example, the
center of a room), a person who has assumed a prede-
termined posture or has made a predetermined motion
(for example, raising his or her hand), a person of the
top priority based on predetermined people's priorities
(for example, within a family, the order of father, mother,
and children) (the priorities can be given based on
weight, sex, age, physiological condition, or physical
and mental handicap), a person who has spoken a par-
ticular word (for example, "yes" or the name of an appa-
ratus), a person having a particular belonging (for
example, a red ball in his or her hand), or a person
directing his or her eyes or head toward a particular
position (for example, an ornament). In addition, the
operator selection section 2 may determine an evalua-
tion value for each person based on an evaluation
method predetermined based on people's attributes in
order to select a person having an evaluation value that
is larger than a reference value and that is also the larg-
est. In this case, the attributes of various people can be
taken into consideration. The evaluation method may
use the weighted sum of the strength of a voice given
when the person says "yes" and the speed at which the
person raises his or her hand.
[0018] Next, the operator selection section 2 presents
information on the selected person. If, for example, the
name of the selected person can be identified, it may be
displayed on a display screen or output as a voice, or
music, sound, or an uttered name associated with the
person beforehand may be output, or a symbol or char-
acters associated with the person beforehand may be
displayed on the display screen, or a signal may be
transmitted to a device carried by the person. The
device carried by the person can provide the information
to the person by vibrating or outputting light or sound.
Alternatively, a light may be directed to the selected per-
son, or a device such as a display may be rotated and
directed to the selected person, or an image of the
selected person photographed by a camera may be dis-
played on the display screen. A voice saying "What do
you want?", or sound or light may be output to only the
selected person immediately after his or her utterance
or motion in order to inform the people of the selected
person.
[0019] Although, in the above example, the operator
selection section 2 selects one from the three people,
several (for example, 2) people may be selected or no
person may be selected if the predetermined attribute
cannot be monitored.
[0020] In FIG. 1 , 3 is the control object candidate
determination section for determining candidates for a
control object based on the predetermined attribute of
the person selected by the operator selection section 2
and his or her peripheral environment. The control
object may include equipment (inside a house, an air
conditioner, a television, a video, a light, a washer, a
personal computer, a game machine, or a pet robot;
outside a house, an elevator or a car) or information or
the contents of display used in information equipment
(characters or graphics displayed on a display screen).
The predetermined attribute of the selected person may
include an indicating posture with his or her finer, gaze,
5 or head, the utterance of particular words, sign lan-
guage, or the holding of a particular article. If the indi-
cating posture with the person's finger, gaze, or head is
used, candidates for a control object will be equipment
near the indicated position or the contents of display on
to the display screen. The peripheral environment may
include temperature, humidity, illuminance, sound vol-
ume, the condition of air currents, the concentration of a
particular gas (carbon dioxide) in the air, or time. If the
indoor temperature or humidity is high, the control
is object may be an air conditioner, a fan, or a dehumidi-
f ier. If the indoor illuminance is low or a particular time
(before sunset) has come, the control object may be
lighting. If the air current does not vary over a long
period of time or the concentration of carbon dioxide in
20 the air exceeds the reference value, the control object
may be a fan or window. If the outdoor sound exceeds
the reference value, the control object may be a televi-
sion or a window.
[0021 ] There may be one or several candidates. If, for
25 example, the person selected as the operator is pointing
toward an air conditioner and a television, both of them
are to be controlled. If the utterance of words or sign
language is used as a predetermined attribute, the
name of the apparatus may be indicated or predeter-
30 mined words may be uttered. For example, the uttered
word "television" allows a television (or televisions) to be
used as a candidate for a control object, and the uttered
word "hot" allows a fan and an air conditioner to be used
as candidates for control objects. If the holding of a par-
35 ticular article is used as a predetermined attribute,
equipment and articles may be mutually associated
before hand. For example, by associating a red ball with
the air conditioner and associating a blue ball with the
television, the holding of the blue ball allows the televi-
40 sion to be used as a candidate for a control object.
[0022] Next, the control object candidate determina-
tion section 3 externally presents information on deter-
mined candidates for control objects. To present
information, a name indicating a control object may be
45 output as a voice or sound or a name for a control object
may be displayed on the display screen. Alternatively, a
light included in an apparatus that is a candidate for a
control object may be turned on or a speaker included in
an apparatus that is a candidate for a control object may
so output a sound or voice.
[0023] In FIG. 1 , the control content candidate deter-
mination section 4 determines candidates for the con-
tent of control based on the predetermined attribute of
the person selected by the operator selection section 2
55 and his or her peripheral environment. The content of
control may include switching on and off equipment,
changing operation parameters (sound volume, wind
quantity or direction, channels, or light quantity) of the
4
:<EP 0919906A2 I >
7
EP0919906 A2
8
equipment, opening and closing a door or window, mov-
ing or modifying an object on the display of information
equipment, changing the color of an object, or editing a
document.
[0024] The predetermined attribute may include the
use of a voice or sign language to indicate the content
of control (for example, the utterance of the word "on"
for switching on, the utterance of the word "off' for
switching off, or the utterance of the word "hot" for
switching on) or predetermined motions may be associ-
ated with the content of control beforehand (clapping
hands once for switching on and clapping hands twice
for switching off). There may be one or several candi-
dates. If the person selected as the operator utters the
word "up", the candidate may be an increase in the vol-
ume of the television or the set temperature for the air
conditioner, or the moving to information of an object
displayed on the display of information equipment
[0025] The peripheral environment may include tem-
perature, humidity, illuminance, sound volume, the con-
dition of air currents, the concentration of a particular
gas (carbon dioxide) in the air, or time. If the indoor tem-
perature or humidity is high, candidates for content of
the control may be the switching-on and increasing
operation force of an air conditioner, a fan, or a dehu-
midifier.
[0026] If the indoor illuminance is low or a particular
time (before sunset) has come, the candidates for the
content of control may be the switching-on of lighting or
increasing light quantity. If the air current does not vary
over a long period of time or the concentration of carbon
dioxide in the air exceeds the reference value, candi-
dates for the content of control may be the switching-on
of the fan or the opening of the window. If the outdoor
sound exceeds the reference value, candidates for the
content of control may be an increase in the sound vol-
ume of the television or the closing of the window.
[0027] Next, the control content candidate determina-
tion section 4 presents information on the determined
candidates for the content of control. To present infor-
mation, a name indicating the content of control may be
output as a voice or sound or displayed on the display
screen.
[0028] In FIG. 1 , the control object and content deter-
mination section 5 determines a control object and the
content of control based on the candidates for a control
object determined by the control object candidate deter-
mination section 3 and the candidates for the content of
control determined by the control content candidate
determination section 4, and then effects the deter-
mined control on the determined control object. The
control object and the content of control are identified by
limiting the candidates for a control object and the con-
tent of control. To limit the candidates, a predetermined
combination of the candidates for a control object and
the content of control is selected. For example, if the
candidates for a control object are the television and air
conditioner and the candidate for the content of control
is to "increase the temperature" and if this content is not
provided for the television but for the air conditioner,
then the control object will be the air conditioner and the
content of control will be to increase the set temperature
5 for the air conditioner. In addition, when as the limitation
of the candidates, the air conditioner is selected as a
candidate for a control object based on the person's
indicating motion and increasing the temperature for the
air conditioner is selected as a candidate for the content
10 of control based on the person's uttered word
"increase", the control object and the content of control
are not selected if the time interval between the time T1
at which the indicating motion was monitored and the
time T2 at which the uttered word "increase" was moni-
15 tored is larger than or equal to the reference value (for
example, three seconds). Likewise, if the time T2 pre-
cedes the time T1 , the control object and the content of
control are not selected. In this manner, the candidates
can be limited by taking the time interval or order
20 between the times T1 and T2 into account. This tech-
nique can reduce the rate of misjudgment caused by a
combination of the accidental occurrence of this indicat-
ing motion and the accidental utterance of this word dur-
ing conversation. In addition, even if the duration of the
25 indicating motion is smaller than the reference value (for
example, equal to or more than 1 second), the control
object is neither selected. This can reduce the rate of
misjudgment if this indicating motion is accidentally
made as a daily motion. This limitation may not be pro-
3 o vided (if, for example, there is only one candidate or
multiple control objects are simultaneously controlled)
when there is no need for limitation.
[0029] Next, the control object and content determina-
tion section 5 presents the information on determined
35 control object and content of control. The determined
control object and content of control can be presented
as in the presentation of the candidates for a control
object or the content of control.
[0030] In addition, the control object and content
40 determination section 5 may indicate confirmation, indi-
cate the needs for reentry, or present information such
as candidate selections and the disability to determine
the control object. With the indication of confirmation,
execution is confirmed for the determined control object
45 and content of control by uttering the word "OK?", dis-
playing it on the display screen, or outputting a predeter-
mined sound. The indication of the needs for reentry
and of the disability to determine the object urge reentry
if the control object and content of control cannot be
so determined easily (for example, the people's predeter-
mined attribute is ambiguous) by uttering the words
"Enter data again", displaying them on the display
screen, or outputting a predetermined sound. With the
candidate selections, if there are multiple control
55 objects and contents of control, the selections are dis-
played on the display screen to urge selection.
[0031] In addition, the monitoring section 1 monitors
the attributes of the operator after the control object and
5
9
EP0 919 906 A2
10
content determination section 5 has presented informa-
tion, and the control object and content determination
section 5 limits the candidates for a control object and
the content of control, determines a control object and
the content of control, and presents information again.
The reduction of the control objects and contents of
control may be repeated based on the presentation of
information and the monitoring of the person's attribute
after presentation. The repetition can reduce the
number of candidates. To confirm execution after the
control object and content determination section 5 has
presented information, the uttered word "Yes" is moni-
tored as the operator's attribute to control the control
object. If the selections of candidates are displayed on
the display screen, they are numbered and the uttered
words indicating these numbers are monitored to con-
trol the control object. The control object determination
section 3 and content determination section 4 may urge
selection from the candidates. In this case, since the
person selects a particular object and content from the
candidates, the control object and content determina-
tion section 5 determines a control object and the con-
tent of control based on this action of selection.
[0032] The operation of the first embodiment of this
invention configured in the above manner is described
below. It is assumed that there are three people A, B,
and C in a room and that there are an air conditioner, a
television, and a fan in this room. The monitoring sec-
tion 1 continuously monitors the people's attributes and
their peripheral environment If A and B point to the tel-
evision, the operator selection section 2 considers B,
who preceded A in pointing to the television, to be an
operator and outputs B's name as a voice. Then, it is
known that A cannot operate the television whereas B
can do it, so this apparatus can be controlled without
confusion despite the presence of the several people.
[0033] In addition to the outputting of B's name as a
voice, the operator selection section 2 can present infor-
mation on the operator by displaying the name of the
selected person on the display screen, outputting
music, sound, or an uttered name as sounds associated
with the person beforehand, displaying on the display
screen a symbol or characters associated with the per-
son beforehand, or transmitting a signal to a device car-
ried by the person. The device carried by the person
can provide the information to the person by vibrating or
outputting light or sound. Alternatively, a light may be
directed to the selected person, or a device such as a
display may be rotated and directed to the selected per-
son, or an image of the selected person photographed
by a camera may be displayed on the display screen. A
voice saying n What do you want?", or sound or light
may be output to only the selected person immediately
after his or her utterance or motion in order to inform the
people of the selected person. By informing the people
of the selected operator in this manner, they know who
can operate the apparatus, and confusion can be
avoided even if several people attempt to operate it.
[0034] The control object candidate determination
section 3 determines candidates for control object
based on the predetermined attribute of the person
selected by the operator selection section 2 or his or her
5 peripheral environment. If B is pointing to the neighbor-
hood of the television and air conditioner, the television
and air conditioner are determined as candidates for a
control object. Even if the air conditioner is away from
the location to which B is pointing, it is included as a
10 candidate for a control object if the room temperature,
which is a peripheral environment, is high. Next, the
control object candidate determination section 3
presents information on the determined candidates for a
control object. The information may be presented by
ys outputting the names of the control objects as voices or
sounds or displaying them on the display screen. Alter-
natively, a light included in the candidate apparatus for
control object may be turned on, or a sound or voice
may be output from a speaker included in the candidate
20 apparatus for a control object. The presentation of the
information on the candidates for a control object ena-
bles the operator to check whether a desired object is
included in the candidates. The selection of the plurality
of candidates enables the desired control object to be
25 reliably included in the candidates.
[0035] The control content candidate determination
section 4 determines candidates for the content of con-
trol based on the predetermined attribute of the person
selected by the operator selection section 2 or his or her
30 peripheral environment. If B says "Strong" while simul-
taneously pointing to a certain location, the candidates
for the content of control will be an increase in wind
force, cooling or warming performance, or illuminance.
[0036] Next, the control content candidate determina-
35 tion section 4 presents information on the determined
candidates for the content of control. The information
may be presented by outputting names for the contents
of control objects as voices or sounds or displaying
them on the display screen. The presentation of the
40 information on the candidates for the content of control
enables the operator to check whether a desired con-
tent is included in the candidates. The selection of the
plurality of candidates enables the desired content of
control to be reliably included in the candidates.
45 [0037] If there are a large number of control objects
and contents of control, the information on all of them
need not be presented.
[0038] The control object and content determination
section 5 determines a control object and the content of
so control based on the candidates for a control object
determined by the control object determination section
3 and the candidates for the content of control deter-
mined by the control content determination section 4,
and then effects the determined control on the deter-
55 mined control object. The method described below can
be used to determine a control object and the content of
control from their candidates. That is, only a predeter-
mined combination of a candidate for a control object
. BNSDOCID: <EP 0919906A2 I >
11
EP 0 919 906 A2
12
and a candidate for the content of control is adopted. In
the above case, if the candidates for a control object are
the television and air conditioner and there are four can-
didates for the content of control, that is, an increase in
wind force, cooling and warming performance, and illu-
minance, the television cannot be combined with the
candidates for the content of control, so the air condi-
tioner is selected as a control object and the content of
control that can be combined with the air conditioner is
limited to an increase in wind force and cooling and
warming performance. Moreover, by monitoring the
temperature or season as the people's environment, the
increase in heating performance can be excluded as the
content of control if the temperature exceeds 30 °C or if
it is in summer. In addition, by recording the history of
the people's control beforehand, the combinations of
control objects and contents of control that have not
been used before can be excluded.
[0039] In this manner, the control object and content
determination section 5 determines a control object and
the content of control based on their candidates and the
peripheral environment, so the control object and con-
tent of control can be more reliably identified even if it is
difficult to identify them separately only from the prede-
termined attributes of people. In particular, if an ambig-
uous attribute of a person such as his or her motion or
posture is used, determinations are difficult even with
very accurate recognition. The present approach, how-
ever, enables a desired control object and the desired
content of control to be selected using the daily motion
or posture of the person and without forcing him or her
to make a dear motion.
[0040] Next, the control object and content determina-
tion section 5 presents the information on determined
control object and content of control. The determined
control object and content of control can be presented
as in the presentation of the candidates for a control
object or the content of control. The presentation of the
information on the control object and the content of con-
trol enables the operator to check whether a desired
control object and the desired content of control have
been selected.
[0041] In addition, the control object and content
determination section 5 may indicate confirmation, indi-
cate the needs for reentry, or present information such
as candidate selections and the disability to determine
the control object. With the indication of confirmation,
execution is confirmed for the determined control object
and content of control by uttering the word "OK?", dis-
playing it on the display screen, or outputting a predeter-
mined sound. The confirmation can prevent erroneous
control. The indication of the needs for reentry and of
the disability to determine the object urge reentry if the
control object and content of control cannot be deter-
mined easily. In this case, reselection is urged by utter-
ing the words "Enter data again", displaying them on the
display screen, or outputting a predetermined sound or
displaying the selections on the display screen. In the
above case, the operator is prompted to aurally indicate
again whether to increase the wind force of the air con-
ditioner or its cooling performance. Then, the monitoring
section 1 monitors the attributes of the operator after the
5 control object and content determination section 5 has
presented information, and the control object and con-
tent determination section 5 limits the candidates for a
control object and the content of control, determines a
control object and the content of control, and presents
10 information again. In the above case, when the person
says "Wind", the control object and content determina-
tion section 5 determines the content of control as an
increase in the wind force of the air conditioner and then
aurally indicates that it will increase the wind force of the
is air conditioner while transmitting a control signal to the
conditioner. The reduction of the control objects and
contents of control may be repeated based on the pres-
entation of information and the monitoring of the peo-
ple's attribute after presentation. The repetition can
20 reduce the number of candidates without misjudg-
merrts.
[0042] In this manner, by indicating the needs for
reentry if the control object or content of control cannot
be identified, misjudgment caused by a forced judgment
25 can be prevented and the equipment or information can
be smoothly controlled using the people's attribute such
as their motion or posture while permitting the ambiguity
of such an attribute.
[0043] Meanwhile in the above embodiments when
30 the reentry from people is monitored, the voice is used
as the people's attribute but another attribute (specific
movement etc.) or combination thereof can be used. For
example when the control object and content of control
which are determined by the control object and content
35 determination section 5 are presented to urge the peo-
ple to determine, people make circle figure by using
arm, hand or finger (see FIG.9(1)-(4)), put up
thumb(see FIG.9(5)) and then the control object and
content determination section 5 determines the pre-
40 sented matter as the control object and content of con-
trol. Further when the people make configuration of "X"
by arm or f inger(see FIG.9(6),(7)),or shake their hand in
a horizontal direction(see FIG.9(8)). The control object
and content determination section 5 does not adopt the
45 presented matter as the control object and content of
control but urge reentry.
[0044] Although in the above example, the pointing
motion or posture is used as the people's attribute in
determining an operator and candidates for a control
so object, other attributes (other i ndicating motions such as
the direction of the eyes or head and a voice, or motions
or postures other than the indicating motions) and their
combinations may be used. If. for example, the people's
position is used as their attribute, a person such as one
55 in the next room who is not related to the operation of
the television is prevented from being mistakenly
selected for an operator. The use of the people's pos-
ture as an attribute, for example, enables a sitting per-
7
QMCnnnrv ,co r>oioon«ao i
13
EP 0 919 906 A2
14
son to be given top priority and prevents the accidental
motion of a person who happens to pass the room from
being mistakenly selected. In addition, using the peo-
ple's face, expressions, identity, or age as their attribute,
those who are limited in the use of the television or who 5
operate the television inappropriately can be excluded
from the candidates for an operator (for example, chil-
dren cannot operate the television after nine p.m.).
Using the people's physical and mental handicap, sex,
or physiological condition as their attribute, an appropri- w
ate person can be given top priority in operation based
on this attribute. Using the people's belonging, an ordi-
nary article can act as a remote controller (for example,
holding a red ball indicates the television), a motion that
the user can use easily or the user's favorable motion 15
can be used to select the television, and the top priority
among several people (who holds the red ball first) for
operating the television can be visibly presented.
[0045] In addition, a combination of multiple attributes
(a voice and the direction of the eyes) enables the 20
equipment to be smoothly operated without forcing the
person to make a particular motion. For example, by
combining the uttered word "television" and the direc-
tion of the eyes to the television, an erroneous reaction
to an accidental motion (the uttered word "television" 25
during conversation) can be prevented.
[0046] The operator can also be appropriately
selected by combining the people's attribute with their
peripheral environment. If, for example, the fan is to be
operated and if the concentration of a particular gas 30
(carbon dioxide) in the air is monitored as the people's
peripheral environment, an appropriate person can be
selected as an operator for the fan based on the peo-
ple's sensitivity to the environment (those who have a
headache when the concentration of carbon dioxide is 35
high). In the operation of the air conditioner, by monitor-
ing the temperature, humidity, or air current condition as
the people's peripheral environment and registering
their sensitivity to the environment (those who are sen-
sitive to the heat or cold and whose skin is likely to be 40
dry in winter) beforehand, a person who is sensitive to
the heat is given top priority in operating the air condi-
tioner if the temperature and humidity are high and if
there are few indoor air currents. If the sound volume of
the television is to be controlled and if the outdoor 45
sound, which is the people's peripheral environment, is
loud, an appropriate person can be selected as an oper-
ator for the television based on the people's sensitivity
to the environment (those who have difficulty in hearing
sound). When the illuminance of the lighting is to be so
changed, the indoor and outdoor illuminances are mon-
itored as the people's peripheral environment to enable
an appropriate person to be selected as an operator for
the lighting based on the people's sensitivity to the envi-
ronment (those who have difficulty in reading characters 55
when dark). If the time is monitored as a peripheral envi-
ronment, the use time can be limited depending on the
person. For example, it is possible to prevent children
from operating the television after nine p.m.
[0047] Likewise, although the above example uses a
voice as the people's attribute in determining the candi-
dates for the content of control, other attributes (other
motions or postures) and their combinations may be
used. For example, by using a motion as the people's
attribute, the equipment can be naturally controlled as in
conversation with the people. Examples of such a
motion include applying the forefinger to the front of the
mouth (see FIG. 10(1)) or plugging the ears (see FIGS.
10(2) and (3)) to reduce the sound volume of the televi-
sion, and keeping the hands near the respective ears
(see FIG. 1 0(4)) to increase the sound volume of the tel-
evision. Further the reproduction of a video etc. is tem-
porally stopped by making such configuration of T" with
using both hands(see FIG. 10(5)) or the switch of TV is
put off by shaking the hand to present "good bye"(see
FIG.10(6)).
[0048] In addition, although, in the above example, the
attribute used to determine the candidates for a control
object differs from the attribute used to determine the
candidates for the content of control, only one attribute
can be used to determine the candidates for both a con-
trol object and the content of control. Examples are
shown below. It is assumed that there are people in a
room and that there are an air conditioner, a television,
a fan, an air cleaner, a telephone, a desk, and a bed in
the room.
[0049] When a person makes a motion of applying
forefinger to the front of the mouth or plugging the ears
with the hands, the control object candidate determina-
tion section 3 determines as a candidate the television
that outputs sound, and the control content candidate
determination section 4 determines the reduction of the
sound volume as a candidate for the content of control.
When a person picks up the telephone receiver, the
control object candidate determination section 3 deter-
mines as a candidate the television that outputs sound,
and the control content candidate determination section
4 determines the reduction of the sound volume as a
candidate for the content of control. When a person
uses the hand to fan the face or body (see FIG. 10(7))
or says "Hot", the control object candidate determina-
tion section 3 determines as a candidate the air condi-
tioner, window or fan that relates to air conditioning, and
the control content candidate determination section 4
determines as candidates for the content of control
switching-on, the reduction of the set temperature, the
opening of the window, the putting on of switch of the
fan, and the increasing the volume of window. Further
people pinch their nose(see FIG. 10(8)) to the control
object candidate determination section 3 determines
the air cleaner or window as the candidate, and the con-
trol content candidate determination section 4 puts on
the switch of the air cleaner or opens the window. When
a person says "good-bye" or "bye-bye" during conversa-
tion, the control object candidate determination section
3 determines the telephone as a candidate, and the
8
:<EP 0919906A2 I >
15
EP 0 919 906 A2
16
control content candidate determination section 4 deter-
mines the disconnection of the telephone line as candi-
dates for the content of control. When a person sits in a
chair and opens a book or holds something to write
with, the control object candidate determination section
3 determines as a candidate the lighting in the room and
the lighting attached to the desk, and the control content
candidate determination section 4 determines switch-
ing-on as candidates for the content of control. When a
person has been sleeping in bed over a specified period
of time, the control object candidate determination sec-
tion 3 determines the lighting in the room as a candi-
date, and the control content candidate determination
section 4 determines switching-off as candidates for the
content of control. Further when people shut a light by
holding up their hand or a member above their eyes(see
FIG.10(9)),or when people indicate such expression as
if light dazzled their eyes, or they say "dazzling" , the
ocntrol object candidate determination section 3 deter-
miens the light source in a room as a candidate and the
control content candidate determination section 4 deter-
mines the making the light dark or switching off as a
candidate.
[0050] As described above, the first embodiment of
this invention enables the people's daily attributes to be
used to control the equipment smoothly without the use
of a remote controller. In addition, even if there are sev-
eral people or apparatuses, this embodiment can
reduce misjudgment and execute control without the
needs for complicate indicating motions.
[0051] FIG. 2 is a block diagram showing a second
embodiment of the invention. In this figure, the configu-
ration of this embodiment and the operation of each
section are the same as in the first embodiment.
According to the second embodiment, however, the
monitoring section 1, operator selection section 2, con-
trol object candidate determination section 3, control
content candidate determination section 4, and control
object and content determination section 5 are attached
to an apparatus to be controlled. If, for example, a tele-
vision incorporates each of these sections, the control
object candidate determination section 3 determines
whether the television is a candidate. Based on the peo-
ple's predetermined attribute, for example, the uttered
words "Switch the channel", the control content candi-
date determination section 4 determines the switching
of the channel as a candidate for the content of control,
and since the switching of the channel is included in the
contents of control for the television, the control object
and content determination section 5 switches the chan-
nel.
[0052] Thus, the second embodiment of this invention
not only provide the same effects as those of the first
embodiment but also only requires the control object
candidate determination section 3 of the apparatus to
determine whether the apparatus has been selected,
thereby reducing the amount of processing required to
determine candidates for a control object compared to
the first embodiment. It is also advantageous that
despite the arbitrary movement of the apparatus, the
processing executed by the apparatus to determine
candidates for a control object does not need to be
5 changed.
[0053] FIG. 3 is a block diagram showing a third
embodiment of this invention. In this figure, the configu-
rations of apparatuses to be controlled 1, 2, and 3 and
the operation of each section are the same as in the
10 equipment to be controlled in the second embodiment
(the control objects 1, 2, and 3 are, for example, a tele-
vision, an air conditioner, and a fan, respectively). The
third embodiment, however, includes multiple appara-
tuses to be controlled. The monitoring section 1 outputs
is monitored contents to a communication network, the
operator selection section 2 outputs information on a
selected attribute of the people to the communication
network, the control object candidate determination
section 3 outputs information on determined candidates
20 for a control object (whether the apparatus to which this
section belongs is a candidate or the degree to which
this apparatus is considered to be a candidate) to the
communication network, the control content candidate
determination section 4 outputs information on the
25 determined content of control to the communication net-
work, and the control object and content determination
section 5 outputs information on the control object and
content of control to the communication network.
[0054] The operator selection section 2 selects an
30 operator based on information obtained from the com-
munication network and information from the monitoring
section 1 in order to present information on the operator,
and the control object candidate determination section
3 determines candidates for a control object based on
35 the information obtained from the communication net-
work and the information from the monitoring section in
order to present information on the candidates for a
control object. The control content candidate determi-
nation section 4 determines candidates for the content
40 of control based on the information obtained from the
communication network and the information from the
monitoring section in order to present information on the
candidates for the content of control, and the control
object and content determination section 5 presents
45 information on the control object and content of control
based on the information obtained from the communica-
tion network and the candidates for a control object
determined by the control object candidate determina-
tion section 3 and the candidates for the content of con-
so trol determined by the control content candidate
determination section 4, and then effects the control on
the control object. This embodiment differs from the
second embodiment in that due to the presence of the
plurality of control apparatuses, control unintended by
55 the operator may be provided when each apparatus
individually determines a control object and the content
of control. For example, if a person points to the inter-
mediate point between the television (a control object 1)
9
PP nQ1QQOAA9 i «►
17
EP 0 919 906 A2
18
and the air conditioner (a control object 2) and says
"Switch on", both the television (the control object 1)
and the air conditioner (the control object 2) are
switched on even if only the television (the control object
1) to be operated.
[0055] In this case, by obtaining, for the other control-
led apparatuses, information on the candidates for a
control object and the content of control and on the con-
trol object and the content of control, the candidates for
a control object and the content of control can be limited
as in the first embodiment. When, for example, the con-
trol object candidate determination section of the televi-
sion (the control object 1) outputs "television" to the
communication network as a candidate for a control
object and the control object candidate determination
section of the air conditioner (the control object 2) out-
puts "air conditioner" to the communication network as
a candidate for a control object, the control object and
content determination section of the television (the con-
trol object 1) determines a control object and presents
the corresponding information based on the information
on candidates for a control object for the television and
the information on candidates for a control object for the
air conditioner obtained through the communication net-
work. In this case, both the television and air conditioner
are candidates, so the television presents information
prompting reentry. The control object and content deter-
mination section of the air conditioner (the control object
2) executes a similar processing. By outputting informa-
tion to the communication network, not only the candi-
dates for a control object, information on the degree to
which the apparatus is considered to be a candidate (for
example, the television: 10; the air conditioner: 5), the
control object and content determination sections of the
controlled apparatuses 1 and 2 compare the degrees to
which the apparatus is considered to be a candidate in
order to determine as a control object the apparatus
having a larger degree (in this case, the television). The
candidates for the content of control is similarly proc-
essed by exchanging information between different con-
trolled apparatuses.
[0056] In addition, by obtaining information on the
operator from the communication network, only the
apparatus to which the operator located closest can
respond without causing multiple apparatuses to
respond to the motion of one person (such as a pointing
motion). If, for example, the operator selected by the
operator selection section of the television (the control
object 1) is identical to the operator selected by the
operator selection section of the air conditioner (the
control object 2) from the communication network, the
above procedure is executed to determine either the tel-
evision (the control object 1) or the air conditioner (the
control object 2) as a control object. In addition, if the
operators are different, the television (the control object
1) and the air conditioner (the control object 2) can exe-
cute processing based only on the attributes of the
respective selected operator. In addition, information
output by the monitoring section of the television (the
control object 1) can be used by the control object can-
didate determination section, control content candidate
determination section, or operator selection section of
s the air conditioner (the control object 2). This has an
effect of providing information that cannot be monitored
using only the air conditioner (information on blind
spots). In addition, information output by the control
object and content determination section of the air con-
w ditioner (the control object 2) can be displayed by the
television (the control object 1). In this case, the output
means (images, voices, or light) of each controlled
apparatus can be shared among the controlled appara-
tuses.
is [0057] The communication network may be wired
(connected via network, telephone, or power lines) or
wireless (using a communication medium such as elec-
tric waves, infrared rays, or supersonic waves), or may
be a mixture of a wired and a wireless networks.
20 [0058] Thus, the third embodiment of this invention
does not only provide the effects of the first and second
embodiments but can also consistently and smoothly
control the equipment based on the people's predeter-
mined attribute even if each controlled apparatus
25 includes a mechanism for individually determining an
operator, a control object, and the content of control.
[0059] FIG. 4 is a block diagram showing a fourth
embodiment of this invention. In this figure, the opera-
tion of each section is similar to that in the first embodi-
30 ment. This embodiment, however, does not include the
operator selection section 2, and determines candi-
dates for a control object and the content of control for
all the people. If, for example, there are N people, the
determination of candidates for a control object and the
35 determination of candidates for the content of control
are executed N times (may be concurrently executed).
The control object and content determination section 5
determines a control object and the content of control
based on the candidates for a control object and the
40 content of control determined for all the people in order
to execute control. The majority rule is used to deter-
mine a control object and the content of control. If, for
example, there are N people and if half or more of the
people issue an instruction for a decrease in the set
45 temperature for the air conditioner, control is provided
such that the temperature for the air conditioner is
reduced. In addition, if the control objects and the con-
tents of control for the respective people are consistent,
the corresponding controls may be simultaneously pro-
50 vided. Thus, the fourth embodiment of this invention
does not only provide the effects of the first embodiment
but can also operate the equipment taking operations
for the several people simultaneously into considera-
tion. To execute similar processing using remote con-
55 trollers, as may remote controllers as the people must
always be prepared that must be operated by each per-
son. This invention enables equipment or information to
be controlled in such a way as to reflect instructions
10
BNSDOCID: <EP 0919906A2 I >
19
EP 0 919 906 A2
20
from several people without the use of a large number of
remote controllers.
[0060] FIG. 5 is a block diagram showing a fifth
embodiment of this invention. In this figure, the opera-
tion of each section is similar to that in the fourth
embodiment. In the fifth embodiment, however, the
monitoring section 1 , the control object candidate deter-
mination section 3, control content determination sec-
tion 4, and control object and content determination
section 5 are attached to a controlled apparatus (for
example, the air conditioner). With this configuration,
the fifth embodiment of this invention does not only pro-
vide the effects of the fourth embodiment but also
requires the control object candidate determination sec-
tion 3 of each apparatus to only determine whether that
apparatus has been selected, thereby reducing the
amount of processing required to determine candidates
for a control object, compared to the fourth embodi-
ment. This embodiment is also advantageous in that
despite the free movement of the position of the equip-
ment, the processing for determining candidates for a
control object for the equipment does not need to be
changed.
[0061] FIG. 6 is a block diagram showing a sixth
embodiment of this invention. In this figure, the opera-
tion of each section is similar to that in the first embodi-
ment. According to the sixth embodiment, however, the
control object candidate determination section 3 deter-
mines candidates for a control object for all the people,
and the control content candidate determination section
4 determines candidates for the content of control for all
the people, tf, for example, there are N people, the
determination of candidates for a control object and the
determination of candidates for the content of control
are executed N times (may be concurrently executed).
[0062] The operator selection section 2 determines an
operator based on the candidates for a control object
and the content of control for the N people. The control
object and content determination section 5 determines
a control object and the content of control based on the
candidates for a control object and the content of control
for the operator selected by the operator selection sec-
tion 2.
[0063] The configuration of the sixth embodiment of
this invention can provide effects similar to those of the
first embodiment. In addition, by selecting an operator
after candidates for a control object and the content of
control have been determined, a procedure based on
the candidates for a control object and the content of
control can be executed by, for example, avoiding
selecting as an operator those for which candidates for
a control object and the content of control cannot be
determined due to the ambiguity of their predetermined
attribute (for example, an indicating motion).
[0064] FIG. 7 is a block diagram showing a seventh
embodiment of this invention. In this figure, the opera-
tion of each section is similar to that in the sixth embod-
iment. In the seventh embodiment, however, the
monitoring section 1, the control object candidate deter-
mination section 3, control content determination sec-
tion 4, operation selection section 2, and control object
and content determination section 5 are attached to a
5 controlled apparatus (for example, the air conditioner).
With this configuration, the seventh embodiment of this
invention does not only provide the effects of the sixth
embodiment but also requires the control object candi-
date determination section 3 of each apparatus to only
10 determine whether that apparatus has been selected,
thereby reducing the amount of processing required to
determine candidates for a control object, compared to
the sixth embodiment. This embodiment is also advan-
tageous in that despite the free movement of the posi-
T5 tion of the equipment, the processing for determining
candidates for a control object for the equipment does
not need to be changed.
[0065] FIG. 8 is a block diagram showing an eighth
embodiment of this invention. In this figure, the opera-
te tion of each section is almost similar to that in the first
embodiment. According to the seventh embodiment,
however, the control object candidate determination
section 3 determines candidates for a control object for
all the people, and the control content candidate deter-
25 mination section 4 determines candidates for the con-
tent of control for all the people. If, for example, there
are N people, the determination of candidates for a con-
trol object and the determination of candidates for the
content of control are executed N times (may be concur-
30 rently executed). The control object and content deter-
mination section 5 determines a control object and the
content of control from the candidates for a control
object and the content of control for the operator
selected by the operator selection section 2. The config-
35 uration shown in the eighth embodiment of this inven-
tion can provide effects similar to those of the first
embodiment.
[0066] Although the first to eighth embodiments have
been described by assuming that the method is imple-
40 merited indoors, they can be adapted for the outdoor
operation of equipment. In addition to the operation of
equipment, these embodiments can be adapted for the
manipulation of information such as manipulation of
objects on a screen or the control of screens.
45 [0067] This invention can be realized using hardware,
or software on a computer, or their mixture.
[0068] This invention is also a program stored medium
for memorizing such programs that are utilized for real-
izing all or part of operations of the control method
so according to the above mentioned present invention.
[0069] As is apparent from the above description, this
invention is a control method for enabling equipment to
be smoothly controlled using people's daily attributes
and without forcing the people to make complicate pre-
ss determined motions, by taking into account the pres-
ence of several people or multiple apparatuses and the
ambiguity of the people's motions and postures that
prevent equipment or information from being smoothly
11
21 EPOS
operated using the people's attributes.
[0070] In addition, a control object and the content of
control can be reliably identified by determining a plural-
ity of candidates for the a control object and the content
of control based on the people's predetermined
attributes and using both information on candidates for
a control object and the content of control to limit the
total number of candidates.
[0071 ] Furthermore, required information can be pro-
vided for the people while a control object and the con-
tent of control can be smoothly determined by
presenting the people with information on the operator
and the content of candidates determined by equip-
ment, and prompting reentry and further observing the
people's attributes if a control object and the content of
control cannot be determined.
[0072] Further according to the present invention
information can be presented by a sound device as well
as by a screen, even if there are a visually handicapped
person and a sound handicapped person, they can con-
firm the information. Therefore the present invention is
useful for welfare work.
[0073] Further when the monitor is executed intermit-
tently, the consume power can be largely reduced.
Therefore the present invention is useful for protection
of the environment of the earth.
Claims
1. A control method for monitoring a person's
attributes and based on the results, in predeter-
mined content of the control, controlling equipment
to be controlled.
2. A control method according to Claim 1 character-
ized by monitoring said person's peripheral environ-
ment and also using these results to execute said
control.
3. A control method according to Claim 1 or 2 charac-
terized by executing said monitoring constantly or
intermittently.
4. A control method according to Claim 1 , 2, or 3 char-
acterized in that based on the results of said moni-
toring, an operator is selected, followed by the
selection of an object to be controlled and the con-
tent of the control.
5. A control method according to Claim 4 character-
ized in that if said control object, said operator or,
said content of control is not determined, such
result information that it is not determined is pre-
sented.
6. A control method according to Claim 5 character-
ized in that at least one of light, characters, voice,
sound, or vibration is used to present the result
19 906 A2 22
information.
7. A control method according to Claim 1 , 2, or 3 char-
acterized in that based on the results of said moni-
5 toring. candidates for an object to be controlled and
for the content of the control are first selected, and
in that an operator is then selected, followed by the
determination of an object to be controlled and the
content of the control.
10
8. A control method according to Claim 1 , 2, or 3 char-
acterized by monitoring several people at a time.
9. A control method according to Claim 1, 2, 3, or 4
75 characterized in that said equipment includes multi-
ple apparatuses.
10. A control method according to any of Claims 1 to 9
characterized by presenting said people with infor-
20 mation on the results of said monitoring and execut-
ing said control based on responses and
instructions issued by said people in response to
the presentation.
25 11. A control method according to Claim 10 character-
ized in that said presented information is at least
one of information on the candidates for said peo-
ple, information on the candidates for said object
equipment, and information on the content of the
30 control.
1 2. A control method according to Claim 1 , 2, or 3 char-
acterized by simply presenting said people with
information of the results of said monitoring while
35 automatically executing said control.
1 3. A control method according to Claim 1 , 2, or 3 char-
acterized in that information on the results of said
monitoring is not presented to said people while
40 automatically executing said control.
14. A control method according to Claim 12 or 13 char-
acterized in that as information on the results of
said monitoring, a plurality of candidates are pro-
45 vided for at least one of the people, equipment, and
content of control, and in that based on the candi-
dates, said control is automatically executed.
15. A control method according to Claim 1 , 2, or 3 char-
so acterized in that a control means for executing said
control is integrated into the object equipment.
16. A control method according to Claim 15 character-
ized in that a said control means is integrated into
55 each of a plurality of apparatuses, and in that the
plurality of control means exchange various infor-
mation using a communication path so that said
control is executed based on the results of the infor-
12
: <EP 0919906A2 I >
23
EP 0 919 906 A2
24
mation exchange.
17. A control method according to Claim 16 character-
ized in that said control means executes said con-
trol by transmitting the results of its own monitoring
to the other control means while adding the con-
tents obtained from the other control means
through said communication path.
18. A control method according to any of Claims 1 to 1 7
characterized in that the attributes of said people
are their motions.
1 9. A control method according to any of Claims 1 to 1 8
characterized in that said control comprises chang-
ing control parameters for the object equipment or
turning them on and off.
20. A control method according to Claim 1 character-
ized in that as the people's predetermined
attributes, the method detects a motion in which
they apply their forefinger to the front of their mouth,
in that the object equipment outputs sound based
on said motion, and in that as a change in the value
of a control parameter, the volume of said equip-
ment is reduced.
21. A control method according to Claim 1 character-
ized in that as the people's predetermined
attributes, the method detects a motion in which
they plug their ears, in that the object equipment
includes equipment outputting sound based on said
motion and equipment controlling the volume
based on said motion, and in that as a change in
the value of a control parameter, the volume of said
equipment is reduced.
22. A control method according to Claim 1 character-
ized in that as the people's predetermined
attributes, the method detects a motion in which
they keep their hand next to their ear, in that the
object equipment includes equipment outputting
sound based on said motion and equipment con-
trolling the volume based on said motion, and in
that as a change in the value of a control parameter,
the volume of said equipment is increased.
23. A control method according to Claim 1 character-
ized in that as the people's predetermined
attributes, the method detects a motion in which
they use their hand to fan their face or part of their
body, in that the object equipment includes equip-
ment controlling air conditioning based on said
motion and a window, and in that as a change in the
value of a control parameter, the volume of said
equipment is reduced.
24. A control method according to Claim 1 character-
ized by continuously or intermittently monitoring the
attributes of a single or plural people, determining
candidates for a control object and the content of
control based on the detection of a people's prede-
5 termined attribute, and determining a control object
and the content of control based on said candidates
for a control object and the content of control.
25. A control method according to Claim 1 character-
w ized by continuously or intermittently monitoring the
attributes of a single or plural people and their
peripheral environment, determining candidates for
a control object and the content of control based on
the detection of a people's predetermined attribute
is and their peripheral environment, and determining
a control object and the content of control based on
said candidates for a control object and the content
of control.
20 26. A control method according to Claim 25 character-
ized in that the peripheral environment comprises
at least one of temperature, humidity, illuminance,
air current, time, or sound.
25 27. A control method according to any of Claims 24 to
26 characterized by outputting to a communication
network, information obtained through said moni-
toring, and/or outputting to the communication net-
work, candidates for a control object and the
30 content of control, and/or outputting to the commu-
nication network, a control object and the content of
control.
28. A control method according to Claim 27 character-
35 ized by continuously or intermittently monitoring
said communication network, and determining can-
didates for a control object and the content of con-
trol based on information from the communication
network and the detection of a people's predeter-
40 mined attribute.
29. A control method according to Claim 27 or 28 char-
acterized by continuously or intermittently monitor-
ing said communication network, and determining a
45 control object and the content of control based on
information from the communication network and
the candidates for a control object and the content
of control.
so 30. A control method according to any of Claims 24 to
29 characterized in that said candidates for a con-
trol object are used as control objects and in that
the candidates for the content of control are used
as the content of control.
55
31. A control method according to any of Claims 24 to
30 characterized in that based on the detection of
the attributes of several people, candidates for a
13
25
EP 0 919 906 A2
26
control object and the content of control are deter-
mined for each of said people, and in that based on
said candidates for a control object and the content
of control, a control object and the content of control
are determined.
32. A control method according to Claim 31 character-
ized in that using the candidates for a control object
and the content of control determined for each per-
son, a control object and the content of control are
determined based on the majority rule.
33. A control method according to any of Claims 24 to
30 characterized in that based on the detection of a
people's predetermined attribute, an operator and
candidates for a control object and the content of
control are determined, and in that based on said
operator and said candidates for a control object
and the content of control, a control object and the
content of control are determined.
34. A control method according to Claim 33 character-
ized in that based on the detection of the attributes
of several people, candidates for a control object
and the content of control are determined for each
of said people, in that an operator is selected based
on said candidates for a control object and the con-
tent of control, and in that based on the attributes of
the selected operator, candidates for a control
object and the content of control are determined.
35. A control method according to any of Claims 24 to
30 characterized in that an operator is selected
based on the detection of a people's predetermined
attribute, and in that based on the attributes of the
selected operator, candidates for a control object
and the content of control are determined.
36. A control method according to any of Claims 33 to
35 characterized in that information on the selected
operator is output to a communication network.
37. A control method according to any of Claims 27 to
29 and 36 characterized in that the communication
network comprises a wireless or a wired communi-
cation network.
38. A control method according to Claims 33 to 36
characterized in that the selected operator is pre-
sented as information.
39. A control method according to any of Claims 24 to
38 characterized in that information is presented
based on the candidates for a control object and the
content of control.
40. A control method according to Claim 39 character-
ized by presenting at least one of candidates for a
control object, a control object, candidates for the
content of control, the content of control, the indica-
tion of confirmation, the indication of reentry, the
selections of the candidates, or inability to deter-
5 mine an object.
41. A control method according to any of Claims 38 to
40 characterized in that at least one of light, charac-
ters, voice, sound, or vibration is used to present
10 information.
42. A control method according to any of Claims 38 to
41 characterized in that based on the results of the
monitoring of the people's attribute after the pres-
15 entation of information, the candidates for a control
object are reduced and/or the candidates for the
content of control are reduced and/or a control
object is determined and/or the content of control
are determined and/or information is presented.
20
43. A control method according to any of Claims 24 to
42 characterized in that based on predetermined
references, the candidates for a control object are
reduced and/or the candidates for the content of
25 control are reduced and/or a control object is deter-
mined and/or the content of control are determined.
44. A control method according to Claim 43 character-
ized in that based on the coherency of candidates
30 for a control object and the candidates for the con-
tent of control, the candidates for a control object
are reduced and/or the candidates for the content
of control are reduced and/or a control object is
determined and/or the content of control are deter-
35 mined.
45. A control method according to Claim 43 or 44 char-
acterized in that using at least one of the time inter-
val between a first time at which a predetermined
40 attribute used to determine candidates for a control
object is detected and a second time at which a pre-
determined attribute used to determine candidates
for the content of control object is detected; the
temporal order of said first and second times; or the
45 duration of the detection of said predetermined
attribute, the candidates for a control object are
reduced and/or the candidates for the content of
control are reduced and/or a control object is deter-
mined and/or the content of control are determined.
50
46. A control method according to Claims 24 to 45
characterized in that the content of control are to
operate the equipment and/or to manipulate the
information.
55
47. A control method according to any of Claims 24 to
46 characterized in that at least one of people's
positions, postures, expressions, motions, voices,
14
BNSDOCID: <EP 0919906A2 I >
27
EP 0 919 906 A2
28
physiological conditions, identities, forms, weights,
sexes, ages, physical and mental handicaps, or
belongings is used as the people's attribute.
48. A control method according to Claims 24 to 47
characterized in that the people's attribute used to
determine candidates for a control object differs
from the people's attribute used to determine candi-
dates for the content of control.
49. A control method according to Claims 24 to 47
characterized in that the people's attribute used to
determine candidates for a control object is the
same as that used to determine candidates for the
content of control.
50. A control method according to Claim 49 character-
ized in that as the people's attribute, the method
detects a motion in which they apply their forefinger
to the front of their mouth, in that the candidates for
a control object include equipment outputting sound
based on said motion and equipment controlling the
volume based on said motion, and in that the candi-
dates for the content of control include control for
reducing the volume.
51. A control method according to Claim 49 character-
ized in that as the people's attribute, the method
detects a motion in which they plug their ears, in
that the candidates for a control object include
equipment outputting sound based on said motion
and equipment controlling the volume based on
said motion, and in that the candidates for the con-
tent of control include control for reducing the vol-
ume.
52. A control method according to Claim 49 character-
ized in that as the people's predetermined
attributes, the method detects a motion in which
they use their hand to fan their face or part of their
body, in that the candidates for a control object
include equipment controlling air conditioning
based on said motion and a window, and in that the
candidates for the content of control include the
reduction of a set temperature or ventilation.
53. A control method according to Claim 49 character-
ized in that as the people's attribute, the method
detects a motion in which they Keep their hand next
to their ear, in that the candidates for a control
object include equipment outputting sound based
on said motion and equipment controlling the vol-
ume based on said motion, and in that the candi-
dates for the content of control include control for
increasing the volume.
54. A program stored medium for memorizing such pro-
grams that are utilized for realizing all or part of
operations of the control method according to any
one of claims 1 to 53.
5
10
15
20
25
30
35
40
45
50
15
EP 0 919 906 A2
i
People
Present
information
>
.<<<<<<<<<
>>>>>>>>>>>>
<<<<<<<<<<<<<
>>>>>>>>>•* • ■ *
)><>( Jment ,<„
<
<
c < < < <
> > > >
c < < *
> >
Control
section
Monitoring
section 1
operator
selection
section 2
Control
object
candidate
determination
sectAog 3
Control
content
candidate
determination
section 4
J
Control object
and content
determination
section 5
Content of
control
Controlled!
equipmenl
vw* — -
16
BNSOOCID: <EP 0919906A2 I >
EP 0 919 906 A2
People
.<<<<<<<<<
>>>>>>>>>>>>
<<<<<<<<<<<<<
Environ-
Present
information
U u
Controlled
equipment:
Control section*
X attached to thelfi MMHHB^H
controlled equip^ entj|«|^
Monitoring
section 1
„ V
Operator
selection
section 2
V
Control
object
candidate
determination
section 3
Control object
and content
determination
section 5
Control
content
candidate
determination
section 4
17
EP 0 919 906 A2
§
-P
O
-p
c
0)
CO
cd
u
04
CD -P
»H 0)
o g
-P-H
C a
O CT
O CD
0)
o
C -P
O G
•«-4 O
-P o
09
i-4 O
O -P
o o
-p
-p
(0
-p
c
CD
a
cr
r— 1
c
o
o
-H
•H
-P
«P
O
o
0)
0)
CD
€U
CO
c
CM
•H
U
H
O
c
O
AJ
o
-P
03
-H
-H
•P
CJ
CD
O
o
Ou
a>
O
0Q
T3
CD
f-l
r-H
c
O
o
M
•H
-P
•P
C
o
o
<D
o
CO
CO
S3
he
-p
C
O
-p
•H
C
s
-P
cu
^5
•H
O
-P
53
CD
-H
D 1
X
w
CD
§
o
•H
-P
-P
O
U
CD
CD
CO
•-I
CD
CQ
C
CM
•H
M
M
O C
O
-P o
-P
ft) *H
•H
M -P
C
CD O
o
04 CD
32
O CO
CD
i-H
r-4
ction
jontro
CD «
CM
CO CD
ent
•h a
it
•P.r*
cr
CD
Communication network
CD CD
o cu
-P S3
-P CT
at cd
on
ion
ecti
lect:
CD
CD
CO
CM
o
c
■p
o
o
<a
-p
M
-p
-H
CD
C3
s
O
CO
-p c
CD (d o
-P C-H
, 4-> <0 -H -P
n o -h C a)
«P -P "5 CD CO
OOflflJC
UO Qt) o
a-
CD A) O
•P CS *H
—1 flJ.H -P
O -P g O
J-l O "H M CD
•P CD *0 CD CO
O JQ <d CD €3
u o o *q_o
a
o
"H
4->
cd c
•p 6
o 5
CD CD
CD
O T3
m
rH
C
O (3
o
M CD
■H
i -p 4J -P
C C
O
O O
CD
O O
CQ
-p
c o
CD M
-P -P
c c
O o
u o
CD
i-H
r— 1
c 2
Ǥ
0> °
CO CD
§^
ent
1-
•H
CT
s-s
CD
18
BNSDOCID: <EP 0919906A2 I >
EP 0 919 906 A2
i S . -4-
people
Present
information
✓ >>>>>/.
.<<<<<<<<<<
+ >>>>>>>>>>>>
<<<<<<<<<<<<<<
>>>>>>>>>>• • • •
<<<<<<<<<< p nv { rn « <
>>>>>>>>>> >r«nviron- >
v>\ J > > V V > > >V > > > > > >
v^S < < ^r^S < < < <
Control
section
> >
< <
>•>
< <
N
> > > >
< < *
> > '
Repeat N times
Monitoring
section 1
Control
object
candidate
determinat ion
section 3 , —
V ■
Control
content
candidate
determination
section 4
J
Control object
and content
determination
section 5
Content
of control
>Contro
equipment!
'EST 1 * ■
19
EP 0 919 906 A2
*>>>>>?.
<<<<<<<<<< .
*>>>>>>>>>>>>
People >>>>>>>>>> • •
>SSSSSSSSS < Environ-S
Present
information
s < n> < > < » < A ment X
vt AvA >,«>v>v
20
EP 0 919 906 A2
People
Present
information
*>>>>>*.
<<<<<<<<<< .
>>>>>>>>>>> >
<<<<<<<<<<<<<<
> > >>>>>>> > E nviiron T
ment < <
>
<<<<<<
> > > > >
< <
> >
< <
N
< < < <
> > > >
Control section
Repeat N times
Monitoring
section 1
Control
object
candidate
determination
v ,
Control
content
candidate
determination
section 4
Operator
selection
section 2
Control object
and content
determination
section 5
Content of control
v
illiisiiii
Control led!
equipment!
21
EP 0 919 906 A2
People
i
*>>>>> + .
. <<<<<<<<<< .
>>>>>>>>>>>>
<<<<<<<<<<<<<<
Present
information
<e <JS^ V? < < y\ ,jL <<<<*< 4
X I > ( 1 > > > > > > > >
22
8NSDOCID:<EP 0919906A2 I >
EP 0 919 906 A2
F i &
People'
Present
information
. < < < <
* > > > >
<<<<<<
> >> > > >
* < < < < < <
>>>>>>
ceo
> > > .
< < < < *
>>>>>>
« < <_ < < < .
>%Environ-
if"V mentV
/■ ^ < < < <
N
Control
section
~T£epea€"
N times
u
> > > >
Monitoring
section 1
3 T
Control
object
candidate
determi-
nation
section 3
Control
content
candidate
determi-
nation
section 4
j
V
Operator
selection
section 2
Control object
and content
determination
section 5
Content of control
23
EP 0 919 906 A2
24
BNSDOCID: <EP 0919906A2 1 >
EP 0 919 906 A2
25
>
(12) INTERNATIONAL APPLICATION PUBLISHED UNDER
(19) World Intellectual Property Organization
International Bureau
(43) International Publication Date
10 October 2002 (10.10.2002)
PCT
(10) International Publication Number
WO 02/080531 A2
(51) International Patent Classification 7 : H04N 5/44
(21) International Application Number: PCT/IB02/00929
(22) International Filing Date: 19 March 2002 (19.03.2002)
(25) Filing Language: English
(26) Publication Language: English
(30) Priority Data:
09/821,183
29 March 2001 (29.03.2001) US
Prof. Holstlaan 6, NL-5656 AA Eindhoven (NL). COL-
MENAREZ, Antonio; Prof. Holstlaan 6, NL-5656 AA
Eindhoven (NL).
(74) Agent: GROENENDAAL, Antonius, W„ M.; Interna-
tionaal Octrooibureau B.V., Prof. Holstlaan 6, NL-5656
AA Eindhoven (NL).
(81) Designated States (national)'. CN, JP, KR.
(84) Designated States (regional): European patent (AT, BE,
CH, CY, DE, DK, ES, FI, FR, GB, GR, IE, IT, LU, MC,
NL, PT, SE, TR).
Published:
— without international search report and to be republished
upon receipt of that report
(71) Applicant: KONINKLIJKE PHILIPS ELECTRON-
ICS N.V. [NL/NL]; Groenewoudseweg 1, NL-5621 BA
Eindhoven (NL).
For two-letter codes and other abbreviations, refer to the "Guid-
(72) Inventors: GUTTA, Srinivas; Prof. Holstlaan 6, anceNotes on Codes and Abbreviations" appearing at the begin-
NL-5656 AA Eindhoven (NL). TRAJKOVIC, Miroslav; ning of each regular issue of the PCT Gazette.
400
C USB EVENT MONITORING^N
^ PROCESS *S
gjg (54) Title: METHOD AND APPARATUS FOR CONTROLLING A MEDIA PLAYER BASED ON USER ACTIVITY
(57) Abstract: A media player controller is disclosed that
^= monitors user activity and automatically controls a media
player in response to predefined events. The disclosed media
player controller includes one or more audio/visual mage
capture devices focused on one or more users. The captured
audio and video information is processed by the media player
controller to identify one or more predefined events. A
number of rules can be utilized to define various user events,
such as when the user has left the room, is on the telephone
or is otherwise paying attention to the media player. Each
rule contains one or more conditions, and, optionally, a
corresponding action-item that should be performed when
the rule is satisfied. Upon detection of a predefined event,
the corresponding action, if any, is performed by the media
player controller.
<
ID
00
o
OBTAIN INPUTS FROM AUDIO/VISUAL
CAPTURE DEVICES) 150
-405
IDENTIFY USa(S) THAT ARE PRESENTT^
410
ANALYZE AUDIO/VISUAL INFORMATION USING AUDIO
AND /OR VIDEO CONTENT ANALYSIS TECHNIQUES
•420
NO,
DOES AUDIO/VIDEO CONTENT
ANALYSIS DETECT A PREDEFINED
EVENT AS DEFINED IN USER EVENT
DATABASE?
1
YES
r
PERFORM ACTION INDICATED IN PROFILE.
IF ANY, OR USER EVENT DATABASE
-440
END
o *-
WO 02/080531 PCT/EB02/00929
Method and apparatus for controlling a media player based on user activity
The present invention relates to methods and apparatus for controlling media
players, and more particularly, to a method and apparatus for automatically controlling a
media player based on user activity.
The consumer marketplace offers a wide variety of electronic devices, such as
5 televisions, stereo systems and personal computers, that provide an ever-growing number of
features intended to increase the convenience and capabilities of these devices. Most
entertainment devices, for example, have an associated remote control device that allows the
user to adjust a number of the device settings remotely. For example, a user can adjust the
program channel, volume and other settings of a television using a remote control, in a well-
1 0 known manner.
While remote controls and other additional features have greatly improved the
convenience of such entertainment devices, they still require the affirmative action of the user
to manipulate the remote control (or another input mechanism associated with the device) to
indicate the manner in which the particular device settings should be adjusted. Thus, if the
15 remote control is not readily available, or the user does not wish to move closer to the device
itself, the user may still be unable to conveniently adjust one or more settings in a desired
manner.
It has been observed that there is often a predictable relationship between
certain user activity and a corresponding manner in which the settings of an electronic device
20 should be adjusted. For example, when the telephone rings while a user is watching
television, the user often responds by manually adjusting the volume or activating the mute
feature of the television. There is currently no mechanism, however, that provides an
indication to an electronic device of such user activity. A need therefore exists for a media
player controller that monitors user activity and automatically adjusts a media player in
25 response to predefined events. A further need exists for a media player controller that
employs a rule-base to define user activities or events, as well as the corresponding response
that should be implemented to adjust device settings.
Generally, a method and apparatus are disclosed for monitoring user activity
and automatically controlling a media player in response to predefined events. The disclosed
BNSDOCID: <WO
02080531 A2 I >
*
WO 02/080531
PCT/IB02/00929
media player controller includes one or more audio/visual capture devices focused on one or
more users. The obtained audio and video information is processed by the media player
controller to identify one or more predefined events.
According to one aspect of the invention, a number of rules define various
5 user activities or events, such as when the user has left the room, is on the telephone or is
otherwise not paying attention to the media player. Each rule contains one or more
conditions, and, optionally, a corresponding action-item that should be performed when the
rule is satisfied to adjust one or more settings of the media player. Upon detection of a
predefined event, the corresponding action, if any, is performed by the media player
10 controller.
A more complete understanding of the present invention, as well as further
features and advantages of the present invention, will be obtained by reference to the
following detailed description and drawings.
15
Fig. 1 illustrates a media player controller in accordance with the present
invention;
Fig. 2 illustrates a sample table from the user profile of Fig. 1 in accordance
with the present invention;
20 Fig. 3 illustrates a sample table from the user event database of Fig. 1 ; and
Fig. 4 is a flow chart describing an exemplary user event monitoring process
embodying principles of the present invention.
25 Fig. 1 illustrates a media player controller 100 in accordance with the present
invention. As shown in Fig. 1, the media player controller 100 includes one or more
audio/visual capture devices 150-1 through 150-N (hereinafter, collectively referred to as
audio/visual capture devices 150) that are focused on one or more user(s) 140 of a media
player 160.
30 Each audio/visual capture device 150 may be embodied, for example, as a
fixed or pan-tilt-zoom (PTZ) camera for capturing image or video information, or one or
more microphones for capturing audio information (or both). The audio and video
information generated by the audio/visual capture devices 150 are processed by the media
player controller 100, in a manner discussed below in conjunction with Fig. 4, to identify one
2
WO 02/080531
PCI7EB02/00929
or more predefined user activities or events. In one implementation, the present invention
employs a user profile 200 and event rules database 300, discussed further below in
conjunction with Figs. 2 and 3, that record a number of user preferences and rules,
respectively. The rules define various events that should initiate an adjustment of one or more
5 settings of the media player 160.
The user activities defined by each rule may be detected by the media player
controller 100 in accordance with the present invention. As discussed further below, each
rule contains one or more criteria that must be satisfied in order for the rule to be triggered,
and, optionally, a corresponding action-item that should be performed by the media player
10 controller 100 to adjust one or more settings of the media player 160 when the predefined
criteria for initiating the rule is satisfied. At least one of the criteria for each rule is a
condition detected in the audio or video information generated by the audio/visual capture
devices 150 using audio or vision-based techniques, in accordance with the present invention.
Upon detection of such a predefined user activity or event, the corresponding
1 5 action, if any, is performed by the media player controller 100. Typically, the corresponding
action is the issuance of a command to the media player 160 to adjust one or more settings.
The commands include, for example, mute, record, volume adjust, change program channel,
power save mode and live pause.
As discussed further below in conjunction with Figs. 2 and 3, the user
20 preferences and rules recorded in the user profile 200 and event rules database 300 may
include one or more criteria that is dependent on external information, such as information
from an optional electronic program guide 130 or caller id (identification) device 170. For
example, the corresponding action-item that is performed by the media player controller 100
in response to a given user activity may be dependent on features of a program, as indicated
25 in the electronic program guide 130. Similarly, the corresponding action-item that is
performed by the media player controller 100 in response to the media player controller 100
detecting that the telephone is ringing may be dependent on the identity of the caller, as
indicated by the caller id device 170.
As shown in Fig. 1, and discussed further below in conjunction with Fig. 4, the
30 media player controller 100 also contains a user event monitoring process 400. Generally, the
user event monitoring process 400 processes the audio information or images obtained by the
audio/visual capture devices 150 and detects one or more events defined in the event rules
database 300.
3
BNSDOCID:<WO 02080531 A2 I >
WO 02/080531 PCT/IB02/00929
The media player controller 100 may be embodied as any computing device,
such as a personal computer or workstation, that contains a processor 120, such as a central
processing unit (CPU), and memory 110, such as RAM and/or ROM. Alternatively, the
media player controller 100 may be embodied as an application specific integrated circuit
5 (ASIC) (not shown) that is included, for example, in a television, set-top terminal or another
electronic device.
Fig. 2 illustrates an exemplary table of the user profile(s) 200 that records
various preferences of each user. As shown in Fig. 2, the user profile 200 is comprised of a
plurality of records, such as records 205-208, each associated with a different user. For each
10 user, the user profile 200 identifies the user in field 250 and the corresponding media
preferences of the user, if any, in field 260.
For example, the user preferences recorded in record 205 for the user, John
Smith, indicates that the user likes to pause the media player 160 when the telephone rings,
unless the call is from a particular telephone number, upon which the volume of the media
1 5 player 160 is lowered. Likewise, the user preferences recorded in record 206 for the user,
Jane Smith, indicates that the user likes to lower the volume of the media player 160 when
the telephone rings, unless the current selected program is a top-5 program, upon which a
record command is sent to the media player 160. Thus, the preferences in record 205 are
dependent upon information from the caller id device 170, and the preferences in record 206
20 are dependent upon information from the electronic program guide 130.
Generally, the user preferences recorded in the user profile(s) 200 can be
obtained explicitly, i.e., from survey responses, or implicitly, by monitoring how a given user
responds to a given set of circumstances. Thereafter, a rule can be established that defines the
given set of circumstances and the correspond action item that should be performed.
25 Fig. 3 illustrates an exemplary table of the event rules database 300 that
records each of the rules that define various user activities or events. Each rule in the event
rules database 300 includes predefined criteria specifying the conditions under which the rule
should be initiated, and, optionally, a corresponding action item that should be triggered
when the criteria associated with the rule is satisfied. Typically, the action item defines one
30 or more adjustments to the settings of the media player 1 60 that should be performed when
the rule is triggered.
As shown in Fig. 3, the exemplary event rules database 300 maintains a
plurality of records, such as records 305-3 1 1, each associated with a different rule. For each
rule, the event rules database 300 identifies the rule criteria in field 350 and the
4
WO 02/080531
PCT7IB02/00929
corresponding action item, if any, in field 360. For example, the rule recorded in record 306
is an event corresponding to the user remaining out of the room (or away from the vicinity of
the media player 160). As indicated in field 350, the rule in record 306 is triggered when the
user remains out of the room for a predefined minimum time interval. As indicated in field
5 360, the corresponding action consists of sending a command to place the media player 160
in a power save mode.
Fig. 4 is a flow chart describing an exemplary user event monitoring process
400. The user event monitoring process 400 processes audio or video information (or both)
obtained from the audio/visual capture devices 150 and detects one or more events defined in
1 0 the event rules database 300. The exemplary user event monitoring process 400 is a general
process illustrating the broad concepts of the present invention. As shown in Fig. 4, the user
event monitoring process 400 initially obtains one or more inputs from the audio/visual
capture devices 150 during step 405. Thereafter, the user event monitoring process 400
optionally identifies the user(s) that are present during step 410, for example, using a
1 5 biometric evaluation of the audio or visual information obtained from the audio/visual
capture device 150. A user identification is particularly useful when the media player
controller 100 permits user-specific media preferences set forth in the user profile(s) 200 to
control over the general rules set forth in the event rules database 300.
Thereafter, the audio/visual information is analyzed during step 420 using
20 audio and/or video content analysis (V CA) techniques. For a detailed discussion of suitable
audio content analysis techniques, see, for example, Silvia Pfeiffer et al., "Automatic Audio
Content Analysis," Proc. ACM Multimedia 96, 21-30, Boston, MA. (Nov. 1996),
incorporated by reference herein. For a detailed discussion of suitable VGA techniques, see,
for example, Nathanael Rota and Monique Thonnat, "Video Sequence Interpretation for
25 Visual Surveillance," in Proc. of the 3d IEEE Int'l Workshop on Visual Surveillance, 59- 67,
Dublin, Ireland (July 1, 2000), and Jonathan Owens and Andrew Hunter, "Application of the
Self-Organizing Map to Trajectory Classification, 5 in Proc. of the 3d IEEE Int'l Workshop on
Visual Surveillance, 77-83, Dublin, Ireland (July 1, 2000), incorporated by reference herein.
Generally, the audio content analysis and VCA techniques are employed to recognize various
30 features in the signals obtained by the audio/visual capture devices 1 50.
A test is performed during step 430 to determine if the audio/video content
analysis detects a predefined event, as defined in the event rules database 300. It is noted that
the general rules set forth in the event rules database 300, as analyzed during step 430, may
be modified in accordance with the specific user preferences set forth in the user profile 200.
BNSDOCID: <WO 0208053 1A2 I
WO 02/080531
PCT/EB02/00929
If it is determined during step 430 that the audio/video content analysis does not detect a
predefined event, then program control returns to step 410 to continue monitoring user
activities in the manner discussed above.
If, however, it is determined during step 430 that the audio/video content
5 analysis detects a predefined event, then the event is processed during step 440 as indicated
in field 260 of the user profile 200, if any, for the identified user or field 360 of the event
rules database 300. Program control then terminates (or returns to step 410 and continues
monitoring user activities in the manner discussed above).
In a further variation, the retention schedule for a given program that is
1 0 recorded in accordance with the present invention can be determined, for example, by a
weight assigned to the program by a user or by a recommendation score assigned by a
program recommender.
A "computer program" is to be understood to mean any software product
stored on a computer-readable medium, such as a floppy-disk, downloadable via a network,
1 5 such as the Internet, or marketable in any other manner.
It is to be understood that the embodiments and variations shown and
described herein are merely illustrative of the principles of this invention and that various
modifications may be implemented by those skilled in the art without departing from the
scope and spirit of the invention.
6
WO 02/080531
PCT/IB02/00929
CLAIMS:
1 . A method for controlling a media player (1 60), comprising
analyzing at least one of audio and video information (150) focused on a user
(140) to identify at least one predefined user activity; and
performing a predefined action item (360) to automatically adjust said media
5 player (160) when said user activity is identified.
2. The method of claim 1, further comprising
establishing at least one rule (305-3 1 1) defining the predefined user activity,
wherein said rule (305-3 1 1) comprises at least one condition (350) and the action item (360)
10 to be performed to automatically adjust said media player (160) when said rule (305-31 1) is
satisfied; and
wherein analyzing at least one of audio and video information (150) is focused
on a user (140) to identify said condition (350).
15 3 . The method of claim 1 or 2, wherein said user activity suggests that said user
(140) is not paying attention to said media player (160) and said action item (360) is the
issuance of at least one of commands to pause said media player (160), or to begin recording,
or to enter a power save mode.
20 4. The method of claim 1 or 2, wherein said user activity is a predefined gestural
command and said action item (360) is the issuance of a corresponding command to said
media player (160).
5. A system for controlling a media player (160), comprising
25 a memory (1 10) for storing computer readable code; and
a processor (120) operatively coupled to said memory (1 10), said processor
(120) configured to
analyze at least one of audio and video information (150) focused on a user
(140) to identify at least one predefined user activity; and
^ BNSDOCID: <WO 02080531 A2 I >
WO 02/080531 PCT/IB02/00929
perform a predefined action item (360) to automatically adjust said media r
player (160) when said user activity is identified.
6. The system of claim 5, wherein the processor (120) is further configured to
5 establish at least one rule (305-3 1 1) defining the predefined user activity,
wherein said rule (305-3 1 1) comprising at least one condition (350) and the action item (360)
to be performed to automatically adjust said media player (160) when said rule (305-31 1) is
satisfied; and
analyze at least one of audio and video information (150) focused on a user
10 (140) to identify said condition (350).
7. The system of claim 5 or 6, wherein said user activity suggests that said user
(140) is not paying attention to said media player (160) and said action item (360) is the
issuance of at least one of commands to pause said media player (160), or to begin recording,
15 or to enter a power save mode.
8. The system of claim 5 or 6, wherein said user activity is a predefined gestural
command and said action item (360) is the issuance of a corresponding command to said
media player (160).
20
9. A computer program product enabling a programmable device when executing
said computer program product to function as the system for controlling a media player (160)
as defined in any of claims 5 to 8.
WO 02/080531
31 *
PCT/IB02/00929
Q >-
II
CD
CC
6t CD
^ CC
S o
->CD
>-
OC
CD CD
CD
CO
co cr»
CD i—
CD
DC
CL.
CO
ggi
OC
LU
CO
CO
LU uj
=! CO
S 2
CO
CO
UJ
CD
1— CD
^ az
LU CL.
Cu CD
OC ^=
LU QC
CO CD
ZD
CD
<---■
6-
Q
LU
or cd
SI
CD
1 ■ I
^ I QC LU
Q ZD | — CD
ZD 92 OL. LU ^
<C>^CD^
V
D^. I CC LU
2;<=5CD
cd zd 1 — =c: _
^ CO LU lo
<><Q
1 I
CD
77
1/3
BNSDOCID: <WO 02080531 A2 I >
WO 02/080531
PCT/IB02/00929
USER PROFILE(S) -200
USER ID
USER MEDIA PREFERENCES
205
JOHN SMITH
SEND LIVE PAUSE COMMAND TO MEDIA PLAYER
UPON PHONE RINGING, UNLESS CALL IS FROM
(777)555-1212, THEN LOWER VOLUME
206
JANE SMITH
SEND LOWER VOLUME COMMAND TO MEDIA
PLAYER UPON PHONE RINGING, UNLESS
SELECTED PROGRAM IS ATOP-5 PROGRAM, THEN
SEND RECORD COMMAND
207
• • •
208
ROBERT SMITH
SEND LIVE PAUSE COMMAND TO MEDIA PLAYER
IF USER LEAVES ROOM;
SEND LIVE PAUSE COMMAND TO MEDIA PLAYER
IF USER IS NOT PAYING ATTENTION TO MEDIA
PRESENTATION
FIG. 2
EVENT RULES DATABASE -300
RULE CRITERIA
350
ACTION
360
305
USER LEAVES ROOM
SEND LIVE PAUSE COMMAND
306
USER REMAINS OUT OF ROOM FOR 1
PREDEFINED TIME
SEND POWER SAVE MODE COMMAND
307
USER IS NOT PAYING ATTENTION TO
MEDIA PRESENTATION
SEND LIVE PAUSE COMMAND
308
USER IS ON TELEPHONE
SEND RECORD COMMAND
309
USER IS SPEAKING TO SOMEONE
ELSE IN ROOM
SEND VOLUME ADJUST COMMAND
310
• • •
311
USER ISSUES PREDEFINED GESTURAL COMMAND
TO CHANGE A SETTING OF MEDIA PLAYER
SEND CORRESPONDING COMMAND
FIG. 3
2/3
WO 02/080531
PCT/EB02/00929
OBTAIN INPUTS FROM AUDIO/VISUAL
CAPTURE DEVICE(S) 150
f
IDENTIFY USER(S)1
fHAT ARE PRESENT
V
405
410
ANALYZE AUDIO/VISUAL INFORMATION USING AUDIO
AND /OR VIDEO CONTENT ANALYSIS TECHNIQUES
420
NO
DOES AUDIO/VIDEO CONTENT
ANALYSIS DETECT A PREDEFINED
EVENT AS DEFINED IN USER EVENT
DATABASE?
YES
PERFORM ACTION INDICATED IN PROFILE,
IF ANY, OR USER EVENT DATABASE
•440
END
FIG,
. 3/3
BNSOOCIDkWO 02080531A2 I >